datamllab/autovideo

AutoVideo: An Automated Video Action Recognition System

55
/ 100
Established

Provides modular pipeline primitives wrapping video processing, augmentation, and recognition algorithms (TSN, I3D, C3D, STGCN) as configurable Python dictionaries, enabling flexible composition and automated search. Built on the D3M infrastructure, it supports end-to-end workflows from raw video ingestion through action classification, with integrated hyperparameter tuning via Ray Tune and Hyperopt. Extensible architecture allows easy addition of new video processing modules while maintaining compatibility with both RGB and skeleton-based action recognition tasks.

340 stars and 62 monthly downloads. No commits in the last 6 months. Available on PyPI.

Stale 6m No Dependents
Maintenance 0 / 25
Adoption 14 / 25
Maturity 25 / 25
Community 16 / 25

How are scores calculated?

Stars

340

Forks

37

Language

Python

License

MIT

Last pushed

Jun 22, 2023

Monthly downloads

62

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/datamllab/autovideo"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.